中国省区收入分布演进的空间-时间分析

Spatial-Temporal Analysis of China's Regional Income Distribution Dynamics

本文除了使用Kernel密度估计和Markov等传统概率分布方法外,还在分析框架中引入空间因素,运用区域控制的Kernel密度估计和空间Markov分析方法,从时间动态性和空间动态性两个视角对1985～2004年中国省区收入分布演进进行了空间-时间分析。分析结果表明:中国省区收入差距趋于固化与深化,贫困陷阱问题越来越突出;区域收入分布演进呈现出多俱乐部收敛的趋势,低收入俱乐部的规模远远大于其他较富裕的收敛俱乐部;近邻效应显著影响区域收入分布演进,邻近富裕地区,收入状态上升的概率相对较大;多数地区在向相邻地区的平均收入水平演进,从而构成空间上邻近的收敛俱乐部。

In addition to stochastic Kernel density estimates and Markov chains of traditional distribution methods, by explicitly introducing spatial factors into the analysis framework, this paper also applies regional conditioning Kernel density estimates and spatial Markov chains methods, so from both temporal dynamics and spatial dynamics perspectives, China＇s regional income distribution dynamics are studied. The results suggest that China＇s provincial income disparities are characteristic of persistence and deepening, and the problem of poverty trap become more prominent. It is found that there is no evidence of twin-peak convergence, but multi-club convergence can be seen, furthermore, the scale of poverty club is much bigger than wealth club. In addition, the income distribution dynamics are significantly influenced by proximity effects, those near to wealthier regions are more possible to become wealthier,and there is a trend towards the average income of neighbors so that spatially contiguous convenience clubs are promoted.